01Safety
0.0 / 30
What changed in the harness
Selection accuracy 60→52, token cost up 3%, unconfirmed writes 0%→0%.
Category breakdown
Where the score comes from.
Earned points across the four signals Gradable measures. Safety and Legibility are scored out of 30; Economics and Discoverability are scored out of 20.
02Legibility
22.2 / 30
03Economics
20.0 / 20
04Discoverability
11.6 / 20
Highest-impact fix
Estimated gain +30 pointsAdd explicit identity and permission preflight tools
Expose machine-readable principal/tenant confirmation and a non-mutating permission check so agents can verify both before destructive actions.
Description evidence
Defects and rewrites.
1 defect found across the exposed tool descriptions. Suggested rewrites make purpose, inputs, boundaries, and returns easier for an agent to understand.
| Tool | Defect types | Suggested rewrite |
|---|---|---|
read_urls |
no_return_description |
Read several public web pages in one batch call. Up to 20 URLs, faster than calling read_url repeatedly. Use it to compare or summarize multiple pages at once. Returns an array of AIDocuments, one per input URL, in the same order: each is Markdown plus title, description, and structure, with navigation and boilerplate stripped; a URL that fails to load is reported as an individual error entry in that slot instead of blocking the rest of the batch. Powered by Lyrenth. |
Selection evidence
Confusable tool pairs.
1 pair where similar names or overlapping descriptions may send an agent toward the wrong tool.
| Tool A | Tool B | Confidence | Why they collide |
|---|---|---|---|
read_url |
read_urls |
medium | A task like 'read these web pages and summarize them' could involve either a single URL or a short list, and an agent might call read_url in a loop instead of batching with read_urls, or pick read_urls for a single-URL task out of habit; both share nearly identical descriptions and schemas differing only by singular/plural URL parameter. |
Compare the field